GenAI, Agents, and the Future of VIS — 2025

In parallel with short paper submissions, the workshop hosted a mini challenge focused on AI agents data visualization. Participants developed agents that automatically analyze a given dataset and generate data reports. The objective was to produce insightful, coherent, and visually compelling analyses of the provided dataset, in the form of a data visualization report.

To support participants, we provided a starter kit based on LangGraph and access to an evaluation server. Participants could also use their favorite framework or library to develop their agent—the only restriction was using the same names for environment variables as specified in the starter kit. By the submission deadline, participants submitted a technical report (VGTC short paper format, up to 2 pages via the PCS system) that (1) describes how the best-performing agent was developed and (2) includes a link to its best generated report. The technical report was archived in the same way as other short papers.

Winners were invited to present at the IEEE VIS 2025 VISxGenAI workshop, and received award certificates and a share of $6,000 in AWS cloud credits: